“Heart disease risk”, “raised cholesterol”, “Statins” – these six words are guaranteed to strike fear into almost anyone who has been unfortunate enough to be cursed by their local witch doctor wielding these hexed mantras. For years the public psyche has been hyper-sensitised to these terms through incessant media reporting and public health messages.
The long discredited image of furred-up pipework haunts many ordinary citizens as they pause momentarily, knives hovering above the butter dish, before reluctantly pushing it away and reaching for the relative sanctuary of the sunflower oil spread, blessed as it is with a little glowing halo of Public Health England approval.
It’s all nonsense of course, but powerful nonsense nonetheless. I don’t intend in this post to stand up against the voodoo of the cholesterol hypothesis and present you with all of the alternative ideas and contradictory data. After all others have done a perfectly good job at that.
Instead I’m going to focus on understanding the current risk charts used by health services across Europe to estimate CVD risk for patients. My aim is to make these risks understandable and meaningful so that the fear is taken out of them, and instead you can use them to make sensible and informed decisions.
So here is the key SCORE Risk Chart that is used in the UK, which is based on the European cardiovascular disease risk assessment model.
To find your risk:
- Identify the column that fits your sex and smoking status.
- Go down to the box that represents your age
- Within that box identify the column for your total cholesterol level – written along the bottom of the chart (4 to 8 mmol/L).
- and find the row for your systolic blood pressure (mmHg) – written up the left hand side of the chart.
The number you find there should be read as a percentage. This is what the GP calls your ‘risk’ of suffering a fatal CVD event in the next ten years.
For example a middle aged (60 year old) non-smoking woman with a cholesterol of 7.2 and blood pressure of 140 would have a 2% risk of dying from CVD in the next 10 years. (I’ll use this example in the remainder of this article)
What risk means
One way of interpreting risk is as if it means chance, as if it were a kind of lottery or Russian roulette where every participant had an equal chance of making the unlucky roll. But this is not what it means.
These figures come from looking at groups of people. A 2% risk actually means that if we watch 100 people who fit in this category (non-smoking woman with a cholesterol of 7.2 and blood pressure of 140) for 10 years then two of them will (on average) die of CVD. In other words what they call risk is really incidence.
The difference between these two ideas is significant. In the former case everyone is at risk. Anyone could get CVD: ‘IT COULD BE YOU’ as the menacing lottery finger points out.
In the latter case, the dice are not necessarily evenly weighted. For example, statistically you would get a 2% death rate for the group as a whole if half the group had a 0% chance, and the other half had a 4% chance. Or if 2 people in one hundred had 100% chance and the remaining 98 had had none.
There are many reasons why each individual may have higher or lower chance of dying from CVD, such as genetics, access to defibrillators, stress levels, comorbidities and even where you live. When giving advice on ‘risk management’ doctors tend to focus on the key features of the chart: stop smoking, reduce blood pressure, get your cholesterol down. But notice that they will also say exercise more. Why? That doesn’t appear in the chart? Neither does obesity or thyroid function or a host of other factors that are at play among the group you have been placed in on this chart.
As an individual you, of course, want to know your individual risk. After all there is no point in getting all worked up if you are at a lower risk than the rest of the group you have been plopped into. Quite rightly, at this point you will be thinking why doesn’t the doctor tell me my individual risk? Why do I need to take on the worry of this say 2% risk when it might not be mine?
The answer, or course, is that at a public health level it doesn’t matter much. If they encourage everyone to stop smoking, prescribe statins and treat the raised blood pressure they will save lives. It’s a statistical certainty. And it’s cheaper than testing all of the other variables that might establish whether you as an individual will benefit or not from their treatment.
▲ Heart disease… The usual suspects. (Image: CQ MSF Films Inc)
Getting to grips with risk reduction
So the doctor is not going to do the testing to establish your individual risk profile, which is little comfort and probably means you are still worried. What can you do? Surely you should take the statins? That’s got to make sense right? Well, that’s your call, but let’s make sure you know what you are doing.
Taking our middle aged woman. If she used statins to lower her cholesterol by 3 mmol/L from 7.2 to 4.2 her risk would shift left three boxes on the table and be 1%. Depending how you spin that it’s either a 1% absolute risk reduction or a much more saleable 50% reduction in relative risk. (Of the two the latter statistical method makes people feel better, especially those with shares in companies producing statins)
You might think this meagre reduction is worthwhile – or simply take the view that any risk reduction is worthwhile. But before you get carried away with the glamour of statistics, lets look at what that means.
Remember that these figures apply to groups not individuals. So what we need to look at is the effects of using statins on the whole group. In which case the reduction from 2% risk to 1% risk really means this:
- If 100 people in the 2% risk group were all placed on statins then over 10 years instead of 2 dying only 1 would die.
- That means that 99 of them took statins for nothing.
- 98 will not die of CVD whether they take statins or not.
- And 1 poor sod will die of CVD either way.
Said another way: there is very little chance that you individually will actually benefit from statins. And the flip side of this coin is that 99% will have to suffer unnecessary medication and any side effects that comes with them.
So at this point I hear the objections being shouted: “But you started with an example who was at very low risk, statins make more sense in high risk patients!”
Fair point. So lets look at a 63 year male smoker with a cholesterol of 7.0 mmol/L and blood pressure of 180 mmHg. He has a whopping 15% risk of dying from CVD in the next 10 years. What should he do?
If he undertook heavy duty statin reduction of cholesterol and dropped it to the lowest category – and did nothing else – he would move from 15% to 10% risk – a 5% absolute risk reduction or 33% relative risk reduction. That would sell it to most people I am sure.
However. If instead he stopped smoking, and got his blood pressure down to 120 mmHg – but left his cholesterol untouched – he would do far better, dropping to a 5% risk. That’s a 10% absolute risk reduction or a 66% relative risk reduction, which is twice as good as the best case for the statins, and gives him all the other health benefits that go with reducing those risk factors. If only he could become a woman he could halve the risk again!
On a personal level the biggest risk reductions matter most. Moving from 15% to 5% at the group level makes a difference to 10 people. If you are in that group there is quite a high chance that you might be one of those lucky 10. But once you have got down to 5% the intensive statin treatment for cholesterol lowering can at best reduce the risk to 3%. All that medication will only benefit 2 in 100. For 98% taking statins at this stage will make no difference to their outcome. 3 will die with or without them, and 95 will survive anyway.
Statins are not even that effective
According to Heart UK, statins typically only reduce cholesterol levels by 29%. So depending on your initial cholesterol level statins will only typically move you 1 or 2 squares to the left. That means for example, that for our smoking male with high blood pressure, statins might be expected to lower his cholesterol from 7.0 to 5.0 mol/L, rather than to 4.0 where the doctor would like him to be.
From what we have looked at so far, the priorities for reducing CVD risk based on the European charts – and hence the emphasis from your doctor – goes like this:
- Stop smoking
- Get your blood pressure down
- Go on statins
From this you might assume that diet comes in forth place. Not so. In fact research is increasingly showing that diet can have a bigger impact on CVD deaths than statins. The PREDIMED trial [Estruch, 2013] for example, found that including nuts or olive oil in the diet and eating closer to the mediterranean dietary pattern decreased CVD deaths by 36% compared to the typical low fat diet. What is more striking is that these results were achieved without restricting calorie intake, and with a higher intake of fats.
Role of diet
As we saw earlier the SCORE risk charts only cover basic variables such as sex, age smoking blood pressure and cholesterol levels. They do, however, come with a range of caveats in the small print which are worthy of attention:
What most of these additional risk factors have in common is their link with diet. The problem for the SCORE risk analysis is that research looking at the effects of diet on these additional small print risk factors often point in the opposite direction to the cholesterol reducing message that fits with the statin story. This is an embarrassment for the standard cholesterol hypothesis and is quietly swept under the carpet to keep things simple. Indeed the authorities make sure that most GPs never have to worry their pretty little heads about these messy dietary details so don’t expect any help in this direction from your GP.
The standard cholesterol lowering advice is as we all know, the low fat diet. However whilst low fat diets do reduce total cholesterol (which ought to reduce CVD risk if the charts are to be believed) they affect other CVD risk factors adversely. In other words lowering your total cholesterol may reduce your risk according to the chart, but increase it in several of the off-chart high-risk categories. For example:
Low fat (high carb) diets have been shown to do the following. [Tang, 2007]
|Effect of a low fat, high carb diet on selected CVD parameters||Expected effect on CVD risk|
|Reduce LDL and total cholesterol||Reduced risk|
|Raises triglycerides||Increased risk|
|Lower HDL cholesterol||Increased risk|
|Increased fasting insulin||Increased risk|
|Increased central obesity||Increased risk|
|Increased BMI||Increased risk|
|Increased fasting glucose||Increased risk|
On the other hand a low-carb (high-fat) diet can reduce CVD risk factors by improving HDL cholesterol and triglycerides, and is effective at managing type 2 diabetes which is otherwise associated with a huge increase in CVD [Meng, 2017]
Just this month MedPage Today published an article on the effects of overwork on the heart. Excessive and prolonged work without proper downtime can cause CVD events and death. The article also mentions “broken heart syndrome”, where sudden emotional stress, such as receiving bad news, results in acute weakening of the heart muscle leading to heart failure or arrhythmias. In a separate article this month they report on evidence that high sugar consumption can raise triglycerides, so should be viewed as a CVD risk.
Another elephant in the room (and wow! there are lots of elephants in this room. I think we are in the elephant house at Cholesterol Zoo) is cholesterol particle size: something the doctors don’t test for. Small, dense, oxidised LDL cholesterol is more atherogenic (risky) than large, buoyant, unoxidised LDL cholesterol. So if your cholesterol is high, but it’s predominantly the latter particle size you have a much lower risk than if your total cholesterol is low, but composed of the small nasty oxidised kind.
This graph can help us appreciate the relative risk caused by excess weight. Being obese (blue bars) roughly doubles your risk.
Earlier I explained that the 10 year risk identified in the chart applies to everyone with the same sex, smoking status, cholesterol and blood pressure, leaving the individual with a wide uncertainty about their personal risk within that group. The additional risk factors that appear in the small print of the SCORE charts provide us with clues about some of this additional variability. The good news is most of these are modifiable through diet and lifestyle. Such changes affect the individual’s risk profile in ways that are not taken into account by the standard risk charts.
When you look at all of the factors that feed in to heart disease, and appreciate the relatively small contribution that total cholesterol makes to the overall picture you have to ask why we concentrate such a disproportionate amount of fear and concern on this one aspect. I suggest that it’s a ploy that suits the medical profession, because along with raised blood pressure, it’s one of the few factors they can control through medication. It’s not surprising then that doctors focus on it so much. However, if you take command of your diet and sort out your work-life balance then you might feel able to drop some of the worry associated with your cholesterol levels.
Giving the game away
To understand how big a part these other factors play consider this. The SCORE charts come in two versions. One for low risk European countries (Andorra, Austria, Belgium, Cyprus, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Israel, Italy, Luxembourg, Malta, Monaco, The Netherlands, Norway, Portugal, San Marino, Slovenia, Spain,
Sweden, Switzerland and the United Kingdom). The other chart (below) is for the remaining European countries, which are classed as high risk (in fact Armenia, Azerbaijan, Belarus, Bulgaria, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Macedonia, Moldova, Russia, Ukraine and Uzbekistan are considered even higher risk still!)
Here is the chart for high risk European countries.
It means a lot of scrolling, but it is worth comparing it to the first chart at the top of this post. Taking our middle age woman and male smoker, if they both moved to one of these high risk European countries their risks would rise from 2% to 3% and 15% to 28% respectively. Isn’t that amazing? Just moving country would undo all the wonderful good those statins were fixing
The first thing to note is that these differences in risk happen without any changes in cholesterol, showing that there are many more factors at play than cholesterol. But, secondly, it should make you stop and wonder how meaningful it is to allocate a risk to an individual in the first place.
Keep Calm and Eat Real Food
The long and the short of it is this. Understand that the ‘risk’ the doctor worries about is statistical. He is worried that you fall into a particular risk category. He doesn’t know what your individual risk is, and he doesn’t really care. His job is to produce results at the group level. Getting you to take his medicine will help his statistics, but not necessarily help you.
Understand what is going on and set about getting your diet right: real food, lower carb. But ignore his advice on diet, he’s only had half an hour training on nutrition in his life! If you follow this blog, or do a little cursory research, you will quickly learn a whole lot more about diet and health than he has even imagined.
In the next post I’ll look at a huge study that investigated cardiovascular disease and diet across Europe – and turned up some interesting results!